Beyond Stereo
Adding to the complexity, Boucher was not only dealing with huge track counts and session material from all over the planet, but today’s surround‑sound requirements for the re‑recording mixer. For Encanto he delivered stems to David Fluhr, who was tasked with mixing all music, singing and dialogue to picture.
“I started the final mixes at the John Ford mixing stage at Fox. Mike flew out to LA for that, and we finished the mixes together, and got them approved by the directors, Lin, and music executives. It is during this stage that I spend the most time on getting the final sounds I want and add most plug‑ins. Everything is in the box, because things change all the time, and there’s no room for recall.
“As I mentioned earlier, we did not overuse plug‑ins. With regards to the vocals, once you’ve lived with their level of dryness, it starts to become the sound of the mix. I’m not going to add a bunch of effects. Sometimes Lin will have some effects in his demos that we like, and I’ll try and imitate those. But in general, it remains pretty dry. When I do add reverb, I lately like to use the LiquidSonics [reverbs], for example Cinematic Rooms, Lustrous Plates and Seventh Heaven. I have a Bricasti, and I know all my presets on that, and you can just put that right into Seventh Heaven, and it’s close.
“I deliver these film mixes to the re‑recording mixer as composite mixes, which are the sum of the stems. When you zero all stems, they are the same as the final composite mix and will phase‑cancel out. For some films I deliver straight 5.1 stems, but that method creates big track counts very quickly when there are on‑screen performances of multiple vocals. So I tend to move to a format of varying track widths.
“On ‘Bruno’ there are 12 lead vocal stems, and four group vocals stems, so that alone is 68 tracks, because the lead vocals are LCR, and the group vocals are 5.0. The instrument tracks are in a variety of formats, like stereo for some of the programmed drums, because they originated from a stereo or mono keyboard, all the way up to 5.1 for the orchestral elements.
“The instrument stems are all in folders. Every stem is a routing folder. As I mentioned, I will print my composites mix from the multitracks, and I will commit to those folders, which are now my stems. The drums are LCR and LFE, ie. low‑frequency extension. Pro Tools does not have a 3.1 format, which is a pain because you need to have an LCR plus a mono LFE.”
Mixing The Soundtrack
After Boucher has delivered his film stems, he creates the stereo and Atmos mixes for the soundtrack album. “Once my film stems are approved, I can make the album mixes, knowing that this is the shape these songs are going to take in the film. That is when I crash them down to stereo. But I don’t just blend the stems. I still have access to all the multitrack tracks. I don’t reference the movie mixes for my stereo mix, because I don’t have the final dub mix yet. I also don’t deal any more with reverbs and compression, and other effects, because the stereo mix needs to be very similar to the film mix. This is based on what I call the MacDougall rule, which refers to Tom MacDougall, President of Walt Disney Music. He wants people to remember the film when they listen to the soundtrack.
“So I can’t turn a song that was in the film and that the audience got attached to into a pop track, as these tend to be over‑compressed, unnatural‑sounding and Auto‑Tuned. I don’t try to make it hit any harder either, because I am already trying to make my mixes hit as hard as I can in the film mix. To push them harder in the soundtrack world would divorce them from the picture.
“I may change minor things in the soundtrack mix. Sometimes I do a little bit more riding and sometimes I add a little more compression. Usually I try to enhance the reverb that is already there, just to make it a little more supportive. By the time I get to the stereo mix, I have seen the movie, and there are moments that suggest that I might make something louder or softer. Then I try to make it stick together more like a stereo mix, as opposed to the full dynamic range of a movie theatre. In a movie theatre you have a captive audience and an enormous dynamic range, but I need to make it sound a little more contained, so it will work in a car or on a set of headphones.
When the time comes to create the stereo mix for the soundtrack album, David Boucher will add a small amount of mix‑bus processing.
“One of the main changes I make on the stereo mix is the stereo bus treatment. I don’t really need much, because I have done the amount of processing that I want to the stems already. The stereo bus chain adds a little extra glue. The stereo bus chain in Encanto has the Acustica Sand2Bus4 [SSL‑style compressor] and the Dangerous Bax EQ. This is very similar to the hardware stereo bus chain in my own studio. There’s also a FabFilter Pro‑L 2, but there’s no gain on it. It’s there for some mild peak‑limiting and to monitor the integrated level.
“After that I do an Atmos mix for spatial audio for the consumer, on streaming platforms, and for Dolby Atmos playback of the soundtrack. Sometimes I use stems, sometimes I use multitracks, sometimes a combination. It depends on what it needs.”
Vocal Recording
Engineer Shawn Coleman meticulously noted his mic placements and gear settings when recording Dianne Guerrero, in case further vocal sessions were needed.
Vocal recording is perhaps the single most important part of the production process, and in the case of Encanto, the pandemic added an extra layer of complexity, as David Boucher recalls: “When a singer is in Colombia, or somewhere else in the world, it takes more time to schedule things. And in general it’s hard to feel collaborative when everybody is remote! So I was really happy to do the Los Angeles sessions in person. The singer and I would both test the day before, and we made sure we had the test results before we met. I did all the recordings with Stephanie Beatriz, the actor for Mirabel, at a dialogue studio in Glendale under the Disney umbrella.
“I go to these vocals sessions with a couple of different mics, and it turned out that my AKG C24 stereo microphone worked best for everyone. Once you pick a mic for someone in an animation production, you stay with it, in case you need to re‑record the singer later. You want to keep the same tone. I also bring my Undertone Audio MPDI‑4 mic pre, and Chiswick Reach valve compressor. The Undertone sounds fantastic, and the compressor also is really beautiful‑sounding on vocals.
“I like using the C24 because the idea is to have a main pickup and a backup microphone in case the main microphone overloads. Especially with dialogue, the dynamics are extreme, and sometimes they are extreme in movie songs as well, because singers can go from a very quiet whisper to almost shouting. I never liked the sound of a more distant microphone, and prefer to record close up, so I used my C24 with one capsule as my main pickup, and the other capsule is padded down, and for backup.”
Boucher also supervised the remote sessions, with suggestions for what mics to use. “Before the remote sessions I was sent video or audio examples of the actors singing, and I looked at the microphone list of each studio where they were going to record, and picked what I wanted as primary mic. For example, at Reservoir Studios in New York they have a great Neumann U67 that we used a lot.
“There was a buddy of mine, Shawn Coleman, who recorded Dianne Guerrero in Atlanta with a Neumann M149. I needed people who could record a singing lead vocal, not dialogue, which is a different art form. After the session Shawn sent me pictures of the mic height, using a tape measure, and pictures of the settings of the gear, so I could match the tone if new vocals had to be recorded with Dianne. I later did a session with her in LA, and put the 149 at 57 inches from the floor and the same outboard settings, and it matched perfectly. The safety mic in this case was a Neumann KM84 with a pad on it.
“Finding people who can really capture a vocalist’s personality was one of the biggest challenges. You can’t just throw up a microphone and set it so it doesn’t clip, because you’re unlikely to get the juice. It’s why I used engineers like Shawn and Joey Raia and Kevin Harp. There also were two engineers in Bogotá, Colombia, at Fox, who did a great job recording.
“The more we could take off the plate of the recordist, the better, because I was desperate to keep them focused on getting a great vocal sound, and not on the technical side. For this reason we sent Shawn and the other recording engineers a Pro Tools session for them to record into. These sessions had a stem of the instruments, a stem of the drums, and all the engineers had to do was set things up for Audiomovers, and adapt the I/O for the place where they were recording.”